Tumors are unstable populations which may progress to increasing malignancy. Acquisition of drug resistance has been related to mutation or amplification of specific genes, but genetic changes which lead to increased tumor growth or to metastasis are generally unknown. The association of malignant properties with expression of certain oncogenes is inconsistent, perhaps because gene expression is determined as an average for the population whereas malignant properties may depend on a small subpopulation. This project will investigate mechanisms of tumor progression. It will examine the relationship between expression of the Ha-ras oncogene or of the Epidermal Growth Factor Receptor gene, and tumor and metastasis formation in immune-deprived mice by cells of human bladder or breast cancer origin. The current strategy of selecting cell lines with different metastic potential and examining for gene expression will be complemented by studying the phenotype of cells after introduction of genetic material, using a marker which will allow selection of transfected cells with high gene content and expression. Further studies of tumor progression are stimulated by evidence for communication between cells, and by the ability to transfer genetic material to living cells in tissue culture. Transfer of genetic information between cells may well occur in vivo especially if there is close contact between cells and high rate of cell death with release of chromatin, as occurs in solid tumors. The potential for horizontal transmission of genetically based properties will be examined making particular use of spheroids as a model for solid tumors. Transfer of repetetive human DNA sequences from MGH-U1 human bladder cancer cells to murine cells will be sought, using selection of murine cells in ouabain. The possible transfer of transforming sequences from MGH-U1 cells to non-malignant murine cells (Balb/3T3 or C3H 10T1/2) will be studied, using the endpoint of tumor formation in syngeneic mice; tumors which arise will be examined for the presence of Ha-ras or other human oncogens sequences. Horizontal transfer may also be an important mechanism for dissemination of drug resistance, and the horizontal transfer of genetically-based, dominant drug-resistant phenotypes will be quantitated. These studies will allow assessment of the role of spontaneous gene transfer as a factor in malignant progression and resistance to therapy.